This invention relates generally to an improved apparatus and method for the performance of chemical processes and, more particularly, to an improved apparatus for automatically performing the sequential degradation of protein or peptide chains containing a large number of amino acid units for purposes of determining the sequence of those units.
The linear sequence of the amino acid units in proteins and peptides is of considerable interest as an aid to understanding their biological functions and ultimately synthesizing compounds performing the same functions. Although a variety of techniques have been used to determine the linear order of amino acids, probably the most successful is known as the Edman Process. Various forms of the Edman Process and apparatuses for automatically performing the processes are described in the following publications:
Edman and Begg, "A Protein Sequenator," European J. Biochem. 1 (1967) 80-91; Wittman-Liebold, "Amino Acid Sequence Studies of Ten Ribosomal Proteins of Escherichia coli with an Improved Sequenator Equipped with an Automatic Conversion Device," Hoppe-Seyler's Z. Physiol. Chem. 354, 1415 (1973); Wittmann-Liebold et al., "A Device Coupled to a Modified Sequenator for the Automated Conversion of Anilinothiazolinones into PTH Amino Acids," Analytical Biochemistry 75, 621 (1976); U.S. Pat. No. 3,959,307 issued to Wittmann-Liebold and Graffunder on May 25, 1976, for "Method to Determine Automatically the Sequence of Amino Acids;" Hunkapiller and Hood, "Direct Microsequence Analysis of Polypeptides Using an Improved Sequenator, A Nonprotein Carrier (Polybrene), and High Pressure Liquid Chromatography," Biochemistry 2124 (1978); Laursen, R. A. Eur. J.Biochem.20 (1971); Wachter, E., Machleidt, H. Hofner, H., and Otto, J., FEBS Lett.35, 97 (1973); U.S. Pat. No. 3,725,010 issued to Penhasi on Apr. 3, 1973, for "Apparatus for Automatically Performing Chemical Processes;" U.S. Pat. No. 3,717,436 issued to Penhasi et al. on Feb. 20, 1973, for "Process for the Sequential Degradation of Peptide Chains;" U.S. Pat. No. 3,892,531 issued to Gilbert on July 1, 1975, for "Apparatus for Sequencing Peptides and Proteins;" U.S. Pat. No. 4,065,412 issued to Dreyer on Dec. 27, 1977, for "Peptide or Protein Sequencing Method and Apparatus." A further apparatus of note is described in copending U.S. patent application Ser. No. 106,828 filed Dec. 26, 1979 by Leroy E. Hood and Michael W. Hunkapiller, two of the applicants hereon, on "Apparatus for the Performance of Chemical Processes."
Briefly, as discussed in the above publications, the Edman sequential degradation processes involve three stages: coupling, cleavage and conversion. In the coupling stage phenylisothiocyanate reacts with the N-terminal .alpha. amino group of the peptide to form the phenylthiocarbamyl derivative. In the cleavage step anhydrous acid is used to cleave the phenylthiocarbamyl derivative to form the anilinothiazolinone. After extraction of the thiazolinone the residual peptide is ready for the next cycle of coupling and cleavage reactions. Aqueous acid is used to convert the thiazolinone to the phenylthiohydantoin which may be analyzed in an appropriate manner, such as by chromatography.
The automated apparatus of the Penhasi U.S. Pat. No. 3,725,010 as modified in the above-referenced articles of Wittmann-Liebold and the pending patent application Ser. No. 106,828 of Hunkapiller and Hood, relates to an automated sequenator in which the reactions are carried on in a thin film formed on the inside wall of a rotating reaction cell which is commonly known as a "spinning cup" and is located within a closed reaction chamber. Means are provided for introducing and removing controlled amounts of liquid reagents relative to the chamber for reaction with a sample of a protein or peptide in an inert atmosphere. The sample to be analyzed is initially placed in the spinning cup, followed by the sequential introduction and withdrawal of the various reagents and solvents necessary for carrying out the coupling and cleavage reactions. The liquid reagents and solvents themselves form films on the walls of the cup which pass over and interact with the sample film as the cup spins. The reagents dissolve the sample film and perform the coupling and cleavage stages of the Edman process. Upon completion of the coupling and cleavage stages, the reaction chamber is evacuated to remove volatile components of the reagents. Following the post-coupling evacuation, the remaining sample film is extracted with solvent to remove non-volatile components. Following the post-cleavage evacuation, the resulting thiazolinone is extracted from the sample film with solvent and transferred either to a separate flask for conducting the conversion step or to an apparatus for collection and drying of the various fractions. In cases where the conversion process is not performed immediately in a conversion flask, the process may be performed later on a number of fractions simultaneously.
The introduction and withdrawal of fluids relative to the spinning cup has been achieved with fluid conduits passing through a plug which seals an opening in the upper wall of the reaction chamber and depends therefrom to a location within the cup. Fluids are introduced directly into the spinning cup at a point adjacent the bottom thereof, and are withdrawn from an annular groove in the cylindrical interior surface of the cup. The fluid to be withdrawn is forced into the annular groove by centrifugal force when the cup is rotated at a high rate, and is withdrawn through a conduit having an inner end projecting into the groove. This effluent conduit thus acts as a scoop for removing the reaction products and by-products and the extracting solvents.
If the protein or peptide sample in a spinning cup device does not have sufficient mass to form a cohesive film by itself, it is sometimes carried on the inner wall of the cup during the solvent extractions within a relatively thick layer of nonprotein carrier material. The carrier material and the sample are then dissolved in the liquid reagents during the reaction stages to enable the coupling and cleavage reactions to take place. A polymeric quaternary ammonium salt having the chemical composition 1,5- dimethyl-1,5-diazaundecamethylene polymethobromide has been used for this purpose. The carrier must be applied in substantial quantities to securely retain the sample, and the carrier and sample are both dissolved by the liquid reagents to permit reaction between the sample and the reagents.
Although devices of the spinning cup type can provide acceptable experimental results in many cases, they have several disadvantages. For example, the expenses of obtaining a suitably large protein or peptide sample and maintaining an adequate supply of the necessary reagents are quite high, primarily because the reagents used are in liquid form and must be used in substantial quantities. Liquid reagents and solvents tend to separate portions of the sample from the film and wash them from the wall of the cup as they pass, reducing the yield of terminal amino acid units obtained in each successive cycle of the apparatus. The initial quantity of sample must therefore be great enough to insure that sufficient sample will remain through the last cycle to produce useful results. Devices of this type also have rather long cycle times due to the considerable volume of the reaction chamber and the need to repeatedly remove semi-volatile liquid reagents and solvents by vacuum drying the sample therein. In addition, spinning cup sequenators are quite complex and expensive, both to manufacture and repair.
A different type of sequencing device is disclosed in the above-cited Laursen and Wachter papers, wherein the sample is immobilized by covalent linkage to the surfaces of a plurality of small beads. The beads form a porous packing within a reaction column, and the column is flooded with liquid reagents to perform the chemical processes. Because the cleavage reagents used are excellent solvents for proteins and peptides, the covalent linkage must be complete in order to hold the sample in place. However, covalent linkage is difficult to obtain in practice. The packed column is also difficult to wash, and the beads therein tend to disintegrate during use.
There have heretofore been proposed sequencing devices designed to overcome the deficiencies of these apparatuses by containing a sample within a stationary reaction chamber and subjecting the sample to at least one reagent in gas or vapor form. The Gilbert and Dreyer patents cited above disclose two such devices, neither of which operates entirely satisfactorily.
The device of the Gilbert patent provides a closed finger-shaped extension within a reaction chamber for holding a peptide or protein sample at a controlled temperature during sequential exposure to gaseous reagents and solvents. Each time a reagent is introduced, the extension is cooled internally to produce condensation thereon. The extension is then warmed, causing the sample to dissolve in the liquid, and the reaction proceeds. After reacting with the sample, the unwanted semi-volatile chemicals may either be dried from the sample by a combination of heat and a stream of inert gas, or be washed from the extension along with the terminal amino acid by a solvent which is condensed on the extension until it drips therefrom.
In the device and method of the Dreyer patent, a protein or peptide sample is applied to both the inner and outer surfaces of many small macroporous beads within a reaction column by chemical coupling or direct adsorption thereto. Various reagents and solvents are passed sequentially through the packed column in either gaseous or liquid form to produce the desired degradation reactions. The flow of reagents and solvents to the column is controlled by a ten position rotary face seal valve.
Unfortunately, the devices of the Gilbert and Dreyer patents do not provide a sufficiently contamination-free environment to achieve acceptable results through a large number of degradation cycles. For example, it is difficult to efficiently wash the protein or peptide sample in the Gilbert and Dreyer devices. The Gilbert method of washing the sample by condensation of solvent thereon to the point at which solvent drips from the sample would tend to leave traces of the various reaction products on the sample contaminating future chemical reactions. Likewise, the packing used to retain the sample within the reaction column of Dreyer is difficult to wash because the various chemical products must be transported entirely through it and away from the column to avoid contamination. This is not easily done even when large amounts of solvent are used, because the solvent tends to pass through the spaces between the beads rather than through the small pores inside the beads where most of the protein sample is located. The fluid feed lines and flow valves of the Gilbert and Dreyer devices are also difficult to fully evacuate and are prone to trapping chemical residues which can interfere with the intended chemistry of further reaction cycles.
The glass or plastic beads used as packing in the reaction column of Dreyer also have a tendency to disintegrate over a number of degradation cycles, clogging the system to the point at which the passage of fluids therethrough is hindered. It then becomes virtually impossible to wash the system between cycles and the chemistry within the column becomes hopelessly contaminated.
The contamination caused by the several factors described above has a cumulative effect over the duration of a sequential degradation process. The sample and the reagents within the reaction cell thus become more and more contaminated, hindering the desired coupling and cleavage reactions and causing a number of undesired reactions to take place. The yield from each complete cycle of the apparatus is thus decreased and a series of contaminants is introduced into the fractions.
The yield is further decreased by direct loss of the sample due to a variety of reasons, including the disintegration of the packing, solubility of the sample in the flushing solvents, and failure of the sorptive bonds between packing and sample.
While these effects may be overlooked in some cases where large amounts of the protein or peptide sample are available or where the chain has a relatively small number of units, they become devastating in cases where the chain has a very large number of units or only very small amounts of the particular protein or peptide are available. Both of these circumstances are present in the case of interferon, a small protein made in human cells in response to certain viral infections. Interferon has recently caused a great deal of excitement in the world of clinical medicine because it promises to be an effective agent for arresting viral infections and it appears to offer considerable hope as an anti-cancer reagent. Interferon is produced and, accordingly, is available only in very small quantities. Currently, virtually the entire world's production of the two types of human interferon originates in the relatively few world centers that have access to large quantities of human white blood cells (leukocyte interferon) or certain human cells in tissue culture (fibroblast interferon). Because of this limited productive capacity of interferon, it has been difficult to carry out well controlled clinical studies and fundamental analyses of how this molecule functions. To further complicate the picture, interferon is composed of a chain of approximately 150 amino acid units, which must be individually cleaved from the chain for analysis. Contamination losses of the types described above can prevent the sequencing of any but the first few amino acid units of interferon with the very small quantities of the protein available. Beyond the first few cleavage cycles, the small sample can become contaminated to the point at which positive results are unobtainable.
The most sophisticated prior device known to the applicants herein for converting the various thiazolinones cleaved from the sample into the more stable phenylthiohydantoins is the conversion flask described in the above-referenced articles of Wittmann-Liebold, as modified in the co-pending patent application Ser. No. 106,828 of Hunkapiller and Hood. However, applicants have found that this flask suffers from inefficient washing of its inner walls when reagents and solvents are introduced through the appropriate capillary tube. It has been suggested that the reagents and solvents can be delivered with a stream of inert gas to wash the flask walls by splattering the liquid thereon, however, this technique causes an erratic flow of liquid to the flask and makes it very difficult to control the volume of liquid delivered.
Applicants have also found that when the prior conversion flask is scaled down appreciably in size to accommodate lower volumes of liquid, it is difficult to obtain the optimum degree of dispersion of inert gas bubbles within the liquid contents of the flask to agitate the contents during the conversion reaction and evaporate the semi-volatile components thereof. Inert gas introduced to the bottom of the flask for these purposes tends to rise to the surface of the liquid in relatively large bubbles which do not uniformly agitate the liquid and instead promote splattering of the liquid onro the top of the flask.
Therefore, in many applications it is desirable to provide an apparatus for performing chemical processes such as the sequencing of proteins or peptides which operates efficiently and with a minimum of system contamination to enable the maximum number of sequencing cycles to be successfully performed with a very small amount of sample.